tiny-asset-pipeline

tiny-asset-pipeline is a node library for compiling and serving web assets. It features dependency management for JavaScript and CSS assets, as well as Less. It's suitable for multi-page applications. Forget Grunt and Gulp,forge configuration.

Install:

npm install -g tiny-asset-pipeline

Run tests:

npm test

Features:

1. Dependency management for Less(or CSS): support mixture @import of .css and .less. All *.css file is processed same as Less file(merge all @import recursively, compile it by lessc). It use the latest(v2.3.1) Less.js.
2. Dependency management for JavaScript: using require("../subDir/foo.js") to require foo.js. Note it's not a CMD or AMD package manager, it just loads/merge the required Javascript file. Now It is a super simple commonjs package loader(although you can't compare it to browserify or webpack) since version 1.3.7 ! It use UglifyJS to process the javascripts.
3. It integrated with autoprefixer. So you can write your CSS rules without vendor prefixes (in fact, forget about them entirely).
4. Note: both require(for js) and @import(for css and less) are using relative path, such as @import ../plugin/foo.css. Images and fonts also use relative path.
5. It will rename asset(append md5 hash to name,such as foo-1d5a631226eed334.js).
6. All other assets except js/css/less, such as images and fonts, are copied and renamed to deploy directoty. All reference urls of images and fonts in the .less/.css are renamed automatically(such as url(icon-update-3da2da84f7287796.png)).
7. A md5_mapping.json will be generated under the root of destination directoty, you can set the -mappingFile option to config the file path and name.

Tips:

1. All *.js , *.less * and *.css will be processed.
2. For partials files: any file(or directory) start with _or . will be ignorged(does not been processed), such as _sub.js or _partial.less or _child.css or _foobar_directory/.

usage:

1. For development environment , start server(as assets pipeline server, compile .less, concat .css and .js per request ): tiny-asset-pipeline -start [-port 8888] [-root yourAssetsRoot]
2. For production environment , batch process before deploy assets to production env : tiny-asset-pipeline -from sourceDir -to destinationDir [-verbose or -v] . Finally it will generate a md5_mapping.json under the root of destination directoty, you can read the content to find the md5 hash of a special file.

Why not Grunt or Gulp?

1. Because I hate to write many configuration.